Development for LHC@home

The ALICE application follows a similar approach to the Theory application but for now no tasks will be available as the work is still required so that the jobs will run in this environment. An announcement will be made when it is ready to be tested.

The ALICE application has been updated and enabled. There may or may not be a few test jobs for the tasks. It is not worth putting any resource behind this for the moment. Try it out only if you are inquisitive.

Congratulations to zombie67 who ran the first successful test job for ALICE!

One of these sneaked in this morning but threw a 207 (0xcf) EXIT_NO_SUB_TASKS as no jobs were available for it to run.
It started up fine, checked for jobs but didn't find any, shut itself down and tidied up on its way out. All seems to be well.
Output looks much like Theory tasks, with a 630MB VM.

I haven't had 1 finish yet, they all stop at about the 1000-1100 second mark or less ... http://lhcathomedev.cern.ch/vLHCathome-dev/results.php?userid=168

I believe, there are currently no jobs available.
However, the tasks should not produce errors, this is for admin to fix.

Congratulations to zombie67 who ran the first successful test job for ALICE!

Weee! However, I am also getting a lot of errors. They all seem to fail at 7 minutes.

---

Reno, NV
Team: SETI.USA

I am just amazed, how some users waste a lot of resources for computing absolutely nothing.

Maybe admin should post, when jobs are ACTUALLY available,or, even better, have an indicator on the Server Status Page for the presence of jobs.

(Don't get me wrong, users can do, whatever they want, but ......)

See the second post in this thread. There is nothing interesting being run at the moment, just the occasional Hello World job. Don't forget that this is the dev project. Computing resources should be thrown production project, we value brain cycles here.

So, why is it so difficult for you to post, when actual work is available, instead of volunteers wasting their time, waiting to test jobs, as they cannot tell, if there are any other then running empty tasks?

Or disable the tasks, if there aren't any jobs.

Is that little bit of courtesy too much to ask?

Hi Rasputin42,

The dev project is used for testing and experimentation so by default things are broken and the CPU cycles used are not valuable for what is crunched but what we learn. At the beginning of this thread it was mentioned that there are only a few test jobs so it is not worth putting any resources behind it for now. As soon as there are real jobs there will be a clear post.

In your account preferences you can easily select which apps you would like to run. I am changing this all the time as often I will to only test the app that is being updated/improved.

For resources which you only want to crunch with, these should be attached the production project.

Having had a few that have only run a few min, this one lasted an hour, attempting to run 2 jobs but maybe not doing so successfully;

2016-07-17 14:15:31 (3624): Guest Log: [INFO] New Job Starting in
2016-07-17 14:15:31 (3624): Guest Log: [INFO] Condor JobID: 1238088 in
2016-07-17 14:41:23 (3624): Guest Log: [INFO] Job finished in with unknown exit code.
2016-07-17 14:41:23 (3624): Guest Log: [INFO] New Job Starting in
2016-07-17 14:41:23 (3624): Guest Log: [INFO] Condor JobID: 1238403 in
2016-07-17 15:09:50 (3624): Guest Log: [INFO] Job finished in with unknown exit code.
2016-07-17 15:15:11 (3624): Guest Log: [ERROR] Condor exited after 3659s without running a job.
2016-07-17 15:15:11 (3624): Guest Log: [INFO] Shutting Down.
2016-07-17 15:15:11 (3624): VM Completion File Detected.
2016-07-17 15:15:11 (3624): VM Completion Message: Condor exited after 3659s without running a job.

Another one that ran for nearly 2hrs, seemingly attempting 3 jobs but all having the line;

 Job finished in  with unknown exit code.

Note the extra space between "in" and "with" as if there is supposed to be a value in there.
Stuff is being written to other logs but I don't know which bits of that would be of value to look at or collect.
No credits again (not that I'm particularly bothered) but that will need to be addressed before these get released into the wild.

It looks like there is a small problem with the logging. I will look into it ASAP

It looks like there is a small problem with the logging. I will look into it ASAP

Laurence, there are several similar examples in the log snippet I sent you today. I suspect there's a typo somewhere in a variable name.

---

It is running in high priority on my SUN Linux workstation and its two core Opteron 1210, 8 GB RAM. There is also a multicore Theory task and a SETI Beta GPU task on its HD 7770 AMD/ATI board. Good old SUN!.
Tullio

The ALICE jobs are just hello world jobs so it is not really worth running them for now. Task 230951 seemed fine and Task 230811 did not get a job, which is also fine.

The ALICE jobs are just hello world jobs so it is not really worth running them for now.

I wanted to say HELLO to ALICE out of VirtualBox version 5.1.4.

Could you pse upgrade the application to vboxwrapper 26197 and remove the distribution of app_config.xml.

Could you pse upgrade the application to vboxwrapper 26197 and remove the distribution of app_config.xml.

Done

11/23/16 15:39:11 (pid:4115) DockerProc::Detect()
11/23/16 15:39:11 (pid:4115) DOCKER is undefined.
11/23/16 15:39:11 (pid:4115) DockerAPI::detect() failed to detect the Docker version; assuming absent.


Starter.log..

Task hangs at stderr.txt --> " HTCondor ping --->0"

Tested several times---ideas?